User Data Set
The User Data Set is a feature of GLIMPS Malware, available on On Premise deployments, that allows each organization to import its own malicious file samples into the platform. These samples are then used to train the Deep Engine on threats specific to the user’s environment, complementing the default dataset provided by GLIMPS. Since the Deep Engine is the shared core across all GLIMPS Malware products, the enriched detection powered by the User Data Set benefits the entire analysis chain.
How does the User Data Set work?
Users import into GLIMPS Malware the malicious samples they want to include in their custom detection. They can group these samples into families that they define within their own dataset.
Once the dataset is set up, the Deep Engine is retrained within minutes to incorporate these new references. The User Data Set then operates alongside the standard GLIMPS dataset: every analyzed file is compared against both the malware families identified by GLIMPS and the families defined by the user.
What is the operational benefit?
Every organization faces a threat landscape unique to its context. Certain malware families target specific industries, particular infrastructures, or exploit attack chains tied to a specific business environment.
The User Data Set enables detection tailored to this reality. By integrating samples already identified internally (for instance, by a CSIRT team during an incident response) an organization permanently enhances the Deep Engine‘s ability to detect variants of those threats, including modified or recompiled versions that would bypass signature-based engines.
Availability and deployment requirements
The User Data Set is available on On Premise deployments of GLIMPS Malware. Since it directly enriches the Deep Engine, the custom detection capabilities apply across all GLIMPS Malware products connected to the instance: network security, business applications, collaboration platforms, email, endpoints, and investigation through GLIMPS Malware Expert.
Dataset management (sample import, family creation, training initiation) is performed through the Expert interface.
It is important to note that the data is not reused for other clients, nor by GLIMPS.